View Full Version : overclockers.com heatsink recomendatiosn so wrong
Microsoft
10-31-01, 12:28 PM
hit the front page looksed at teh recommnedations and had to laugh.
my zalman 5000 smokes all those heatsinks listed and i dont even have a 8000rpm delta. i should get one though and really trash the heatsinks there
And what temps are you getting. Screen pic please. load and idle.
Thanks
Direwolf
10-31-01, 12:43 PM
give us some pictures
Microsoft,
You are always saying what you have in have in your box blows everything out of the water and not backing it up with hard data.
I am sure if you would show the proof it would be of interest to everybody. But statements without any substance are just a waste of your typing. Don't take me wrong this is not a personal attack. I would just like to see the pics of your temps.
Microsoft
10-31-01, 01:22 PM
well here they are
Those look like great temps. What are you using to get your load temp?
heh, that shut everyone up
FYI. The new Motherboard Monitor also reads your cpu usage.
Microsoft
10-31-01, 01:36 PM
i usually play redfaction, and my temps go up some.
but i have to get some panyhose so i can turn a fan or two back on , dont need dust in my case
Let's take it easy kids. As I'm fond of saying, your mileage may vary. Each system is different as far as case, fans, components, air ventilation and circulation, etc., etc. To be truly scientific (I am a senior member by hobby and a scientist by profession), you need to test and evaluate all those heatsinks on the same system to keep things consistent in order to make a fair comparison. Maybe Microsoft has an excellent system with excellent cooling. That doesn't mean someone else using a different system and different heatsink can't equal or top that cooling potential. I've had wonderful success with my GlobalWin FOP cooler, better than my Alpha PAL6035. Others have claim they run their system cooler with the Alpha compared to the FOP. Ok, that's fine and I believe them. Different systems are different. Let's all be "cool". Peace.
VashTheStampede
10-31-01, 01:38 PM
It doesn't shut me up, I don't believe those temps. Especially not with a "quiet" cooler with a .62 C/W. Your getting between 21 C and 30 C with a Zalman on a 1.4GHz OC'ed to 1.535. Hell the Alpha PAL-8045 with a .21 C/W only gets 47.9 C on a T-bird 1.397. That and when this very site reviewed your cooler, it gave temps in the upper 50s on a 1.2GHz!
A) Those temps are faked.
or
B) The sensor is way off
or
C) You've rigged up some form of air cooling that actually makes the Zalman 5000 look good.
~RT~
I would love to see what your temps are running, prime 95, F@H, or Seti. I am looking into getting a new heatsink since I can't stand the noise of my cak 38, which doesn't even cool that good.
Microsoft
10-31-01, 01:46 PM
im using a 50cfm sunon on a 80 to 60 adapter
if i stick a 8000 rpm 60mm delta pushing 68cfm it will be lower than what is listed on the front page
its ok people dont know how to install a heatsink which i understand you need skill.
Microsoft
10-31-01, 01:47 PM
http://www.tenarrus.com/kcdms/cooler-mod001.gif
I see that you are a SETI supporter. What are your temps while you are running SETI?
Microsoft
10-31-01, 02:13 PM
no i just link it
If I had 2 Sunons blowing on to my chip and my PSU pulling heat off, I'd have low temps too. Nice setup.
But thats not your heatsink doing the job, like you're saying. Its the fans you have blowing onto your heatsink are doing the job.
Wish I could do that w/ my crummy case . . .
killem1x1
10-31-01, 02:34 PM
For the people that stated they didn't believe Microsoft, that simply isn't right. I visit these forums becasaue I trust the people here. I do not expect everyone to prove everything. Having said that, I do enjoy seeing pics of benches, people's systems, mbm screenshots, etc.., but I certainly don't assume someone is lying, especially if they are a member here.
Now, having said that I do see why some of you feel that way. Microsoft sometimes you do come across like an old member that used the name Would71. I don't mean that as a cut or a flame, but you could soften the edges of your posts a little, if you care to ofcourse ;)
This is just my slice of humble pie :)
That Lian Li case is nice, the extra case fans are nice, the 80mm adapter is nice. Like I said in my earlier post, it's a combination of lots of things that can make the difference.
I would still like to know the load temps he gets. and if those fans are sucking or blowing?
Microsoft
10-31-01, 02:41 PM
then people would want me to help them, and i cant have that.
I also know what other sites have shown and know from experiance that people tend to listen to people that dont know what there talking about.
but they all get there followings and if you manipulate them enough they believe you.
thats one reason i **** alot of people off cause i dont fall for there stuff and tell it how it is
robertm
10-31-01, 02:44 PM
I think Microsoft owns stock in Zalman. :)
Microsoft
10-31-01, 03:15 PM
Originally posted by robertm
I think Microsoft owns stock in Zalman. :)
nope i just know quality stuff when i see it
and i have yet to crack a core
and same reason i use epox motherboards.
quality
Cool. Honesty is what we need but with tact. Now to keep things in perspective you got to give all the details in your test. Idle & Load temps, during the test not right after, using a outboard temp checker. Ambient temps during all temp checks, if you could control the environment it would be great, & other well known HSFs. Only then will you be able to say that your tests are to be taken as the truth. Nobody expects anything from you. I try to help people because I want to not 'cause they expect me to. Don't believe everthing you read.
Warlord2
10-31-01, 04:20 PM
I have a NICE watercooling setup with 2 radiators and a maze2 and get worse temps thin that
I guess it could make a difference in where you live but right now if it was 10c cooler in here I would need a jacket
see what happens if you turn off some case fans and leave your case open your temps shouldnt go up that much maybe 2-3c
Wicked Klown
10-31-01, 04:55 PM
how do you find out what your ambient temp is.
WyrmMaster
10-31-01, 06:12 PM
Just so everyone knows, MBM reads my CPU temp 15c cooler than it actually is. Food for thought...
Same board. mmmmmmmmmmmm!
Microsoft
10-31-01, 06:52 PM
Originally posted by WyrmMaster
Just so everyone knows, MBM reads my CPU temp 15c cooler than it actually is. Food for thought...
then you have a bad board. y temps are the same in mbm and the bios
and you have to set then board to the correct sensorwhich is sensor via686b-2
WyrmMaster
10-31-01, 07:23 PM
Originally posted by Microsoft
then you have a bad board. y temps are the same in mbm and the bios
and you have to set then board to the correct sensorwhich is sensor via686b-2
By "cooler than it actually is" i mean than the CPU core's ACTUAL TEMPETURE, measured by a compunurse probe touching the edge of the CPU core. MBM should read the same as the BIOS, they both read the same sensor, the one inside the socket.
Thelemac
10-31-01, 09:16 PM
Temps that are taken using the insocket thermistor are completely irrelavent when you compare to heatsinks. They are completely irrelavent when you compare the same heatsink with different fans. They are completely irrelavent when you compare the same heatsink with the same fan but change the airflow in the case.
Essentially, MBM is only useful in making sure that you aren't about to kill your chip. An alarm system. A safety precaution. If you really want to know how your cool your chip is you're going to have to do some work. There's just no easy way around it.
Here (http://www.overclockers.com/tips507/) is a link to explain this in another way, with a bit more definition and proof.
oc jason
10-31-01, 09:59 PM
Your getting between 21 C and 30 C with a Zalman on a 1.4GHz OC'ed to 1.535. Hell the Alpha PAL-8045 with a .21 C/W only gets 47.9 C on a T-bird 1.397.
__________________________________________________ __
I beleive them, heck according to all progs to monitor temps and also BIOS i get 41c idle and 44c load with a 1.33 @1.53 using a non delta 29cfm fan on a Glaciator
r0ckstarbob
11-01-01, 05:11 AM
check your guages and then go find an alternate thermostat and test it with that. after you're done, reconcile your temperature guage settings and then come back and let us know what you're really getting.
RSB
this reeks of BS.
KILLorBE
11-01-01, 06:23 AM
Did you bend the insocket thermistor up a bit so that it touches your CPU? that could make a huge difference. :o
Also what's wrong with your -12V and -5V ? (both say 0V :p )
Microsoft
11-01-01, 06:29 AM
i have 2 systems withthe 87ka and the other one runs way cool also and its at a dift location
Morphoius
11-01-01, 02:06 PM
Originally posted by Warlord2
I have a NICE watercooling setup with 2 radiators and a maze2 and get worse temps thin that
I guess it could make a difference in where you live but right now if it was 10c cooler in here I would need a jacket
see what happens if you turn off some case fans and leave your case open your temps shouldnt go up that much maybe 2-3c
Watercooling gets your ship much colder than using any fan setup, but because there is no fan on the waterblock warm air builds up around the socket. if you want that thermometer to show lower temps then just point some fans at the socket. that will cut quite a bit off that temp, but not increase your overclockability becuase the actual chip temperature does not change. Whoever posted it was right, temp probes under the CPU are little more than allarm systems.
Like I said before, your temps are so low because you have:
a) A 60mm fan on your sh!tty Zalman
b) 2 92mm Sunons (or a like brand) blowing onto your HS
c) You're power supply fan taking warm air off of your HS
(see first page of thread for pic of this)
Don't act like you have the uber HS or something, because the HS isn't doing the work. Those 4 big fans are doing all of the work for your heatsink. Yeah, if I was moving 300 CFM around my HS, I'd get ungodly low temps as well, but I'm not.
I'm not saying its not an impressive setup, because it is. But you're acting like its your Heatsink that is doing all of the work, when its not really doing much at all.
VashTheStampede
11-01-01, 02:59 PM
Originally posted by Oni
Like I said before, your temps are so low because you have:
a) A 60mm fan on your sh!tty Zalman
b) 2 92mm Sunons (or a like brand) blowing onto your HS
c) You're power supply fan taking warm air off of your HS
(see first page of thread for pic of this)
Don't act like you have the uber HS or something, because the HS isn't doing the work. Those 4 big fans are doing all of the work for your heatsink. Yeah, if I was moving 300 CFM around my HS, I'd get ungodly low temps as well, but I'm not.
I'm not saying its not an impressive setup, because it is. But you're acting like its your Heatsink that is doing all of the work, when its not really doing much at all.
Exactly, my air cooling setup makes the ineffiecient TT Super Mini Orb look good.
4 60mm
2 120mm
1 60mm PSU fan
4 BayCoolers
Now Microsoft, my challenge to you is to restore the original Zalman fan, remove all fans, other than your PSU fan, from your system and retake your temps. What? Scared your going to fry your OC'ed 1.4?
~RT~
You're doing a great job of cooling off your cpu.
TECRIDER
11-01-01, 03:12 PM
Nice Setup!
I am a newbie and am crossing that cooling bridge in setting up my new system and I have to say your pics gave me some better insight on doing this.
Just one Question are you blowing air on the cpu and exhausting out the back with the 2 sunons or VisaVersa.
Could you tell me how they are blowing?
Could you help a new overclocker to be?
I want to bring my temps down before any overclocking.
Thanks for any advice!:)
Thelemac
11-01-01, 03:24 PM
Ok guys, you're all getting just a bit out of hand. We don't tolerate personal attacks here. The last half of this thread has been rather heavily edited.
If you would like to make personal comments, please do so privately. This is not a forum to flame other people.
Thanks
Thelemac
lol, I like your editing of my post thelemac.
Come on, you know everyone is thinking the same thing.
I can at least agree with Microsoft that epox puts out quality boards
Owenator
11-01-01, 03:41 PM
Originally posted by Thelemac
Ok guys, you're all getting just a bit out of hand. We don't tolerate personal attacks here. The last half of this thread has been rather heavily edited.
No Sh..., I mean right on Thelemac!
I don't think we want to berate each other like a bunch of 11 year olds on a play ground! Can't we all just get along!
If you have to curse, make sure you use my favorites like 'Fargin Cork-suckers' or 'Bastages'!
Ok, my bad, sorry! just quite an amusing to read this stuff while I'm at work!
O
WyrmMaster
11-01-01, 04:45 PM
Originally posted by Microsoft
i have 2 systems withthe 87ka and the other one runs way cool also and its at a dift location
Yes, but the point we are all trying to make is that the insocket thermisor is WRONG, it always is, always will be. Makes no difference if the board manufacture makes quailty stuff or not, measureing the temp of the back of the CPU does not tell you the core temp.
VashTheStampede
11-01-01, 04:57 PM
Originally posted by WyrmMaster
Yes, but the point we are all trying to make is that the insocket thermisor is WRONG, it always is, always will be. Makes no difference if the board manufacture makes quailty stuff or not, measureing the temp of the back of the CPU does not tell you the core temp.
And to further elaborate on that point, DansData.com:
Don't trust your motherboard!
If you're monitoring your CPU temperature with some sort of software utility that reads a motherboard sensor, beware. It's better than using a sensor stuck to a heat sink, but it isn't necessarily a lot better.
Socket A motherboards that can monitor CPU temperature usually have a probe behind the CPU itself, in the middle of the socket. If the probe's touching the back of the chip package, it'll pretty much tell you how hot the chip package is.
Unfortunately, the hardware that reads the probe output may or may not be reliable. People's hardware-monitor temperature readings sometimes change drastically when they update the BIOS version of their computer. The temperature, of course, has stayed the same; it's the calibration of the monitoring hardware that's changed.
And many behind-the-chip probes don't quite touch the CPU.
And probe readings are susceptible to changes in ambient temperature that soak through the motherboard.
A better way yet to monitor CPU temperature is with a thermal sensor inside the chip package itself. All current Intel CPUs have a thermal sensor, which better motherboards can read; the new breed of AMD "Athlon 4" processors apparently have an internal sensor as well.
Internal sensors aren't right smack dab in the middle of the CPU core, but they nonetheless give you a quite good idea of the chip temperature. They're still at the mercy of the monitoring hardware calibration, though.
William
11-01-01, 10:29 PM
That is utterly amazing because that would make it the best heatsink ever made. If you calculate everything out you have a .11 C/W better than a Swifty with a delta, an Sk-6, Glaciator, 8045, etc. Heck, that is better than most watercooling setups. Your temps HAVE to be affected by all those fans, no doubt. A .18-.30 C/W i could see, but not that good. I believe that you have a great setup there, it looks excellent and I rather like the design of the Zalman, but there is NO WAY they are that good. In Socket Thermistors are flawed by their nature, no matter how good the motherboard is(I rank Epox as maybe the top right now). AMD says no to using the thermistor and I think they know how to measure the temps of their CPUs.
NuclearFusi0n
11-01-01, 10:40 PM
Originally posted by kendan
I would still like to know the load temps he gets. and if those fans are sucking or blowing?
****
^testing word censors
Silversinksam
11-01-01, 11:23 PM
I guarantee you if you slapped in a Swiftech MC462-A or an ALPHA PAL 8045T to name just two in your current setup you will get better results. No offense to you but Zalmans are only good if you want a quite hsf.
Didnt we have this discussion before microsoft? I know I posted links showing the Zalman being trounced by several heatsinks.
Ps didnt you mention you were getting rid of that Zalman 5000?
holy hell!! a 14k rpm fan?
SickBoy
11-02-01, 09:57 AM
Don't buy it. notice his case temps are 15 C to 18 C. (And dont try to tell me that Sensor2 is your case, you can't cool below ambient with aircooling). I'm sure if I turned my thermostat down to 50 F I could get awesome temps too. Stick a PAL8045 in your system and I bet your temps will drop even further.
(And FYI, the 60MM Delta is available in 38, 44 and 50 CFM varieties. The 80MM is available in 44 and 69 CFM varieties. You seem to be getting them confused)
SickBoy
VashTheStampede
11-02-01, 10:25 AM
Hmm Microsoft seems to be avoiding this thread.
~RT~
Originally posted by RedneckTech
Hmm Microsoft seems to be avoiding this thread.
~RT~ That's 'cause he can't come up with proof.
I see a Kenshin fan in you. Amakakeru Ryu No Hirameki:cool:
With Microsoft's CPU putting out 86.2 watts and a HS C/W of .62, the CPU should be running at 53.4 C over ambient (C/W * watts = Delta T.) It's obvious there is a socket themistor compression factor of between four and five.
NuclearFusi0n
11-05-01, 10:24 AM
http://hardocp.com/reviews/cooling/roundup0601/index5.html
Zalman matched the SK6
Thelemac
11-05-01, 11:39 AM
Originally posted by NuclearFusi0n
http://hardocp.com/reviews/cooling/roundup0601/index5.html
Zalman matched the SK6
You can't trust those results, though, because the test procedure if qutie inaccurate. The use of the insocket thermistor is bad (see previous posts in this thread for why) and the ambient temp is "controlled," which really can't be done. There is no way to control the temperature in an entire room. Since you can't control it, you have to measure it, and incorporate that into your results.
Random Nonsense
11-05-01, 11:54 AM
maybe all those fans are forcing air UNDER the socket and directly cooling the sensor.... another way is that most boards allow adjustment of the sensor... maybe it is set to display 10 celsius under or something like that?
i am fascinated by microsoft insisting the HSF is this good.... anyone else here have one?
Im gonna do some experiments on my new rig to see if i can replicate his results..... wish me luck on a 20 celsius palomino 1.53 :) using a glaciator 2
NuclearFusi0n
11-05-01, 03:36 PM
I have a Zalman CNPS 5000+ and i got it under the recommendation of a friend. I still haven't installed it, it's sitting in a corner looking for a home :D eh, everybody that I know that has one really likes it...
NuclearFusi0n
11-05-01, 03:46 PM
Originally posted by Thelemac
You can't trust those results, though, because the test procedure if qutie inaccurate. The use of the insocket thermistor is bad (see previous posts in this thread for why) and the ambient temp is "controlled," which really can't be done. There is no way to control the temperature in an entire room. Since you can't control it, you have to measure it, and incorporate that into your results.
http://hardocp.com/reviews/cooling/roundup0601/index.html
It's not accurate for predicting temps, but it's quite adequate for comparing two different coolers
Originally posted by NuclearFusi0n
I have a Zalman CNPS 5000+ and i got it under the recommendation of a friend. I still haven't installed it, it's sitting in a corner looking for a home :D eh, everybody that I know that has one really likes it...
What kind of temps do your friends get? load, Idle, what cpu, and a description of their setup would be appreciated.
Thelemac
11-05-01, 04:42 PM
Originally posted by NuclearFusi0n
http://hardocp.com/reviews/cooling/roundup0601/index.html
It's not accurate for predicting temps, but it's quite adequate for comparing two different coolers
Not true. Still too many variable to work with. Even sealing off the socket like that does not eliminate the inaccuracies due to cooling secondary heat paths. He also makes no mention of ambient temps. Just because he did it on different days (only five) does not make it statistically accurate. I *did* read that before hand to see if he had changed anything...found that he did, but not enough to satisfy me. The insocket thermistor is *only* good for comparing the same heatsink with slight changes. It isn't just the airflow under the socket...it's the airflow around it. They are just fine for an alarm, but there is no way you can possibly make that an accurate system to test the differences between coolers.
WyrmMaster
11-05-01, 06:10 PM
Originally posted by NuclearFusi0n
http://hardocp.com/reviews/cooling/roundup0601/index.html
It's not accurate for predicting temps, but it's quite adequate for comparing two different coolers
Read any of the heatsink comparisions of the front page and you will see instances where one two heatsinks differ by 4c by the insocket thermistor and by 12c by a real temp sensor. Look at this one, in the second chart compare the top and bottom rows.http://www.overclockers.com/tips519
NuclearFusi0n
11-07-01, 10:52 AM
Originally posted by WyrmMaster
Read any of the heatsink comparisions of the front page and you will see instances where one two heatsinks differ by 4c by the insocket thermistor and by 12c by a real temp sensor. Look at this one, in the second chart compare the top and bottom rows.http://www.overclockers.com/tips519
The absolute temperature doesn't matter. It's how two heatinks relate that matters.
Thelemac
11-07-01, 12:12 PM
Originally posted by NuclearFusi0n
The absolute temperature doesn't matter. It's how two heatinks relate that matters.
That's isn't really what he was referring to. It was how the differences between MBM and the temp of the back of the CPU don't really seem to have much correlation.
Like Oni said, the forums are correct in their findings.They rated the heatsink.It's your fan doing all the work.Put the same fan as they used in the test setup and see if your temps blow them out then.Also these forums are about helping people out and sharing info not just posting "your" stuff.It should be a give and take thing and your previous response sure makes it seem like your only a taker.People are curious as to how you achieve such great temps and you post them to show them off but then don't want to be bothered with helping someone else achieve these temps.I believe every word of what you posted and think that's great every system is different but I think you should take a different attitude toward these forums and not just take but be willing to explain how you achieve these things so others can prosper also.w
WyrmMaster
11-07-01, 07:12 PM
Its not so much that i dont believe him, i just pointed out that his temp measurement system is inacurate at best. When microsoft comes in and says that our heatsink recomendations are wrong, and his only data comes from a system that everyone knows is fately flawed, well, we dont like it.
krakerman
11-07-01, 10:36 PM
yeah, the idea that that lousy heatsink beats the others is completely wrong... hell, i have the same heatsink and cooling setup getting temps ranging 10c on 2 mobos. that sensor is VERY inaccurate and all i use it for is making sure something doesnt fry. also rednecktech, is that kenshin pic from that awesome fight between kenshin and saitoh? ooOOOo that gives me the shivers ;-)
vBulletin® v3.8.7, Copyright ©2000-2012, vBulletin Solutions, Inc.